In response to the issue of global warming, the world's leading companies are striving towards green manufacturing. By examining the electricity consumption statistics of various plants, it is found that the chiller system in the facilities accounts for an average of 41% of the total facility electricity consumption and 20% of the total plant electricity consumption, making it the most energy-consuming system in the entire plant. In view of this, this project takes the chiller system of the first plant of the world's leading company as a case study. Using AI algorithms, a neural network model is established to calculate the optimized settings for chiller temperature, cooling water temperature difference, and cooling water return temperature. The AI then automatically feeds back and controls these settings, using the most suitable values to enhance overall efficiency and achieve energy-saving goals.